Administration of Resources in System-Wide Search Systems

1. A method for administration of a search system for searching data input from a plurality of applications, the method comprising: receiving from a plurality of applications, indications of intent to send data for input into an index stored in storage of a search system at specified times after the indications, wherein the data for input is to be indexed in the search system, and wherein each indication including: requirements of an application for acceptable performance parameters of the search system for said data for input; and the specified time at which the data for input is intended to be sent; adapting the search system based on said requirements by allocating or de-allocating performance resources of the search system to meet the requirements specified in said indications for a plurality of inputs from different applications at the specified times; including using Optimization techniques to dynamically allocate or de-allocate said performance resources to meet the applications' requirements, when an indication is received, wherein the optimization techniques are selected from the group including: queuing theory, linear programming, task scheduling; and receiving data for input into the index stored in the storage of the search system at the specified times from the plurality of applications.

2. A method as claimed in claim 1 , including defining policies regarding resource consumption in predetermined circumstances.

3. A method as claimed in claim 1 , including reserving said performance resources for an application prior to the input of said data.

4. A method as claimed in claim 1 , including prioritising applications to determine resource availability to the applications.

5. A method as claimed in claim 1 , wherein the requirements of the plurality of applications include one or more of: the relative priority of an application, an expected rate of update of the data, the average size of the data and the data type, a maximum excepted delay between sending data for input to the search system and the data being searchable.

6. A method as claimed in claim 1 , comprising determining that there are sufficient performance resources to meet an indication of an application, and issuing a specification of the performance parameters available to specify a minimum quality of service the application will receive.

7. A method as claimed in claim 1 , including: receiving said indications from a plurality of applications in advance of sending data for input into the search system; adapting the search system before receiving the data for input; and subsequently receiving data for input into the search engine from the plurality of applications.

8. A method as claimed in claim 1 , including defining policies regarding resource consumption at predetermined times.

9. A method as claimed in claim 1 , including indexing the data received from the plurality of application in the search system.

10. A method of providing a service to a customer over a network for administration of a search system for searching data input from a plurality of applications, the service including the steps of: receiving from a plurality of applications, indications of intent to send data for input into an index stored storage of a t-he search system at specified times after the indications, wherein the data for input is to be indexed in the search system, and wherein each indication including: requirements of an application for acceptable performance parameters of the search system for said data for input; and the specified time at which the data for input is intended to be sent; monitoring the performance resources of the search system; and adapting the search system based on said requirements by allocating or de-allocating performance resources of the search system to meet the requirements specified in said indications for a plurality of inputs from different applications at the specified times; and including using Optimization techniques to dynamically allocate or de-allocate said performance resources to meet the applications' requirements, when an indication is received, wherein the optimization techniques are selected from the group including: queuing theory, linear programming, task scheduling.